Sum Worker - Java SDK Sample

This sample demonstrates how to create a simple ArmoniK worker using the Java SDK. The worker receives a name as input and returns a greeting message.

Prerequisites

Java 17 or higher
Maven 3.9+
Docker
A running ArmoniK cluster (see Local Deployment)

Project Structure

sum-worker/
├── pom.xml
└── src/main/java/fr/aneo/armonik/worker/samples/
    ├── SumProcessor.java
    └── Main.java

How It Works

The SumProcessor implements the TaskProcessor interface from the ArmoniK Java Worker SDK:

public class SumProcessor implements TaskProcessor {

  @Override
  public TaskOutcome processTask(TaskContext taskContext) {
    // Read inputs
    var num1 = Integer.parseInt(taskContext.getInput("num1").asString(UTF_8));
    var num2 = Integer.parseInt(taskContext.getInput("num2").asString(UTF_8));

    // Compute result
    var result = num1 + num2;

    // Write output named "result"
    taskContext.getOutput("result").write(String.valueOf(result), UTF_8);

    return TaskOutcome.SUCCESS;
  }
}

The Main class bootstraps the worker server:

public static void main(String[] args) throws IOException, InterruptedException {
    var armoniKWorker = ArmoniKWorker.withTaskProcessor(new HelloWorldProcessor());
    armoniKWorker.start();
    armoniKWorker.blockUntilShutdown();
}

Building the Worker

Build the JAR and Docker Image

./mvnw clean package

This command:

Compiles the Java code
Creates a shaded JAR with all dependencies
Builds a Docker image named sum-worker:latest using Jib

Verify the Docker Image

docker images | grep sum-worker

Expected output:

sum-worker   1.0.0-SNAPSHOT   <image-id>   <date>   <size>
sum-worker   latest           <image-id>   <date>   <size>

Deploying to ArmoniK

Step 1: Deploy ArmoniK Locally

Follow the ArmoniK Local Deployment Guide to set up a local ArmoniK cluster.

Step 2: Create a Partition for the Java Worker

ArmoniK uses partitions to group workers with specific configurations. You need to create a partition that references your sum-worker Docker image.

Edit the partitions.tfvars file in your ArmoniK deployment directory (typically infrastructure/quick-deploy/localhost/):

compute_plane = {
  # partition for sum worker
  sum = {
    replicas    = 0
    socket_type = "tcp"
    polling_agent = {
      limits   = { cpu = "2000m", memory = "2048Mi" }
      requests = { cpu = "50m", memory = "50Mi" }
    }
    worker = [{
      image    = "sum-worker"
      tag      = "latest"
      limits   = { cpu = "1000m", memory = "1024Mi" }
      requests = { cpu = "50m", memory = "50Mi" }
    }]
    hpa = {
      type              = "prometheus"
      polling_interval  = 15
      cooldown_period   = 300
      min_replica_count = 0
      max_replica_count = 5
      behavior = {
        restore_to_original_replica_count = true
        stabilization_window_seconds      = 300
        type                              = "Percent"
        value                             = 100
        period_seconds                    = 15
      }
      triggers = [
        { type = "prometheus", threshold = 2 }
      ]
    }
  },

  # existing partitions
  default = {
    # ... existing default configuration
  }

}

Step 3: Apply the Partition Configuration

Redeploy ArmoniK to apply the new partition:

cd infrastructure/quick-deploy/localhost
make deploy

Step 4: Verify the Partition

Open the ArmoniK Admin GUI in your browser (typically at http://localhost:5000/admin) and navigate to the Partitions section. You should see your helloworld partition listed with its configuration.

Running a Client

To test the worker, use the Java Client sample that submits tasks to the sum partition.

See the Java Client Sample for detailed instructions on how to run a client that interacts with this worker.